Tong Zhang

 

 

Tong Zhang

 

 

 

 

Talk: Reinforcement Learning for Foundation Models: Theory, Algorithms, and Applications 

This talk provides an overview of our group’s research on reinforcement learning (RL) for foundation models. I will discuss how RL principles can improve the training, alignment, and capabilities of large language models, highlighting the theoretical foundations, theoretically motivated algorithms, and their applications to foundation model training, digital assistants, embodied agents, and scientific discovery. I will also outline our ongoing efforts to build more capable, interactive, and reliable agents across diverse domains using RL.

Workshop Home Page

Return to the TAU-UIUC Workshop Home Page

Home→